Understanding the factors that lead individuals to engage in healthy or unhealthy diet and exercise patterns is a necessary first step in the prevention of obesity and other chronic conditions. There is a need for new approaches for understanding why some persons engage in more health-promoting behaviors while others are less engaged in these. This application proposes an interdisciplinary project using a theoretical model based on self-control concepts so as to better understand variation in exercise and diet behaviors. Self-control models posit that aspects such as planning, problem solving, and self-monitoring constitute a set of related abilities that may contribute to health outcomes. However, self-control models have not been extensively applied to prediction of adolescent health behavior, and it is not clear how individual self-control characteristics may interact with environmental factors. This research proposes to use a school-based approach to obtain data on range of self-control measures and study how these measures are associated with body mass index, diet, physical activity and sedentary behavior, and how self-control measures may interact with environmental factors (neighborhood and school characteristics). The research will examine measurements of self-control in the context of other constructs from social cognitive theory. Data will be collected using a school-based survey in a multi-ethnic sample of middle and high school children (7th, 8th, 9 and 10th grades) in the New York Metropolitan area (N = 2,000). The sample of public school students will have approximately 30% African-American, 33% Hispanic, and 24% White non-Hispanic children. We will use hierarchical models to control for potential confounders and account for clustering. Analysis will determine the correlation of self-control measures and environmental measures with diet and exercise behaviors, and the sample size will allow exploratory analyses on how these relations differ by gender and ethnicity. Moderation analyses in hierarchical modeling will test for interaction between environmental variables with individual characteristics. Findings from this research will have implications for the translation of self-control research for the design of prevention programs in school or community settings. [unreadable] [unreadable] [unreadable]